-
PROF. HAROLD E. EDGERTON*
Massachusetts Institute of Technology
Cambridge, Mass.
Supplementary Lighting in
Underwater Photography
Sound will provide the principal way by whichmen and instruments
will do precise navigationin the sea.
ABSTRACT: Photography in the sea, particularly the deep sea,
requires a lightsource near the camera because daylight from the
surface does not penetrate.~onttnuous sources of light require an
energy source which may be large andmvolved, and are, therefore,
only used for visual observation, for motion-picturephot~graphy,
and f~r television. Flash lighting is almost universally
employedfor nngle-puture sttll photography as ample energy is
contained in a small bat-tery to expose thousands of photographs.
Calculations are shown to aid in thedesign of an illumination
system that allows for the absorption and scatteringeffects of sea
water.
SUPPLEMENTARY LIGHTING IN THE SEA
H UMAN OB~ERVATIONS AND photographyat depth In the sea must be
accomplishedwith auxiliary lighting as sunlight is rapidlyabsorbed
with depth. Even at 30 meters belowthe surface, sunlight is very
feeble. A sourceof light should be taken into the sea whenworking
at almost any depth.
If close to the surface and below a ship, anelectrical cable can
be used for power tooperate an over-vol ted tungsten lamp.
Cous-teau has used this system most effectively inhis remarkable
award winning motion pic-tures The Silent World and World
WithoutSun. He uses tungsten lamps that are over-vol ted
momentarily during the operation ofthe motion picture camera. The
diver-photog-rapher signals the engine room of the R/VCalypso by
means of a buzzer when the fullpower of the lamp is required for
photog-raphy. Cousteau uses a daylight type ofcolor film with
tungsten illumination. Thecolor balance of the camera-light is
correctedat one distance because the water absorbs the
* Presented at the Annual Convention of theAmerican Society of
Photogrammetry, Washing-ton, D. c., :vIarch 1967 as one of several
paperson underwater photography, all contained in thisissue.
red light more than the blue. A yellow pictureresults for closer
distances than this and ablue-green one for further. The casu~l
audi-ence attending an underwater movie is notconscious of these
subtle color changes asthey are so occupied by watching the
actionon the screen.
Strobe and flash lamp lighting equipmentare of the greatest
importance in the sea forstill photography, because the
illuminationfrom the sun ceases to be useful at shallowdepths due
to the absorption and scattering inthe water. It is always night or
bad weather orfoggy when one goes below the surface of thesea.
Supplementary lighting equipment is anessential item whenever still
cameras are usedin the sea, not only to furnish light, but also
toovercome the color unbalance, and unevenlighting from light
transmitted from the sur-face.
The expendable flash lamp is of great usefor underwater
photography, because of thelarge quantity of light that is
available fromthe chemical burning of the oxygen and metalfoil. A
slow shutter speed 0/25 sec.) is de-sired so that as much light as
possible can beused. As with the motion pictures, Cousteauand
others use the expendable flash lamp at adistance of 3 meters plus
from the subject and
906
-
SUPPLEMENTARY LIGHTING IN UNDERWATER PHOTOGRAPHY 907
depend on the selective absorption of thewater to reduce the red
light so that thebalance is correct for daylight color film.
Adaylight blue Aash lamp can be used at closerdistances in water,
for example, less thanabou t 2 meters.
Electronic Aash is an advantage under-water if many photographs
are taken becausethe diver does not need to replace the lamp ashe
does with the expendable Aash lamp. An-other important property of
an electronicAash source for underwater photography is itsability
to produce thousands of Aashes effi-ciently from a small electrical
battery. Thus,the inability to control exactly where andwhen a
photograph is to be made underwaterby suspended cameras can be
compensatedpartly by taking many thousands of photos.
The photographer of underwater subjects,especially deep-sea,
seldom has the artisticfreedom of his above-water brother. Most
ofthe time the underwater camera operator isshooting blind with a
camera by some sort ofremote control. It is possible to go with
thecamera in a bathyscaphe or diving saucer as aview finder, but
this is a luxury arrangementwhich is not available for most routine
deep-sea photography assignments. In some cases atelevision camera
can be used as a remoteview finder to aid the photographer to
selectthe right moment to release the shu tter.
Unfortunately, one cannot see or photo-graph greater than
approxi mately 30 metersfrom the camera, even in the clearest of
oceanwater. A more practical limit for photographyseems to be 10 or
15 meters in mid-ocean atthe bottom where conditions are good.
Be-yond 15 meters a haze absorbs light anddestroys the photographic
image. Many at-tempts have been made to overcome theoptical
limitations by selecting the mostfavorable wave length, but to my
knowledgeno spectacular results have been attained.One may conclude
that the camera shouldhave a wide-angle lens which should be keptas
close as possible to the subject, so thatoptical effects from the
water are minimized.Likewise, the ligh ts should be separated
fromthe camera and closer to the su bject than thecamera in order
to prevent back scattering,absorption, and to give shadows and
model-ing.
Photography of the bottom of the ocean isof great importance. To
cover a square milewith 10-by-10-meter photographs will requiresome
29,000 photographs if no photo-overlapis obtained. This is a lot of
photos to process,study, and position. Accurate bottom map-ping by
photography is a big task. There are
-1i(2.
many square miles in the ocean! And thenthere is also the space
between the bottomand the surface to be explored.
A few practical systems of underwaterphotographic equipment will
now be de-scribed and some examples will be shown.Figure 1 shows a
typical arrangement of twocameras (stereo pair), an electronic Aash
unit,and a pinger on a metal rack (Unistrut sys-tem). Dr. J. B.
Hersey, formerly with \\ToodsHole Oceanographic Institute, and now
at theOffice of Naval Research, says "A singlecamera is only one
half of a stereo camera sys-tem!" His experience wi th stereo has
im-pressed him with the added information thatthe stereo
presentation produces, and he in-sists on a two-camera stereo pair.
He alsowan ts color in at least one of the stereo cam-eras, because
it gives added information.
Another underwater camera arrangementused on W.H.O.r. ships is
shown in Figure 2.Note that the strobe is closer to the bottomand
off the axis of the camera as this position-ing tends to red uce
the back scatter from thestrobe. A si milar arrangemen t with as
manyas three cameras and two 200-watt-secondstrobes have been used
to obtain photographswith the cameras at 15 meters above thebottom
in clear ocean water. The Unistrutsystem enables a camera user to
modify or
-
908 PHOTOGRAMMETRTC ENGI EERING
change his camera-lamp arrangement quicklyto suit his whims.
An improved camera system, with 400 footreels of film instead of
100, has been devel-oped in response to a request from
'vV.H.O.I.and others who wish to take many photos inthe sea.
The recen t discovery of very large sharks atgreat depths (1100
fathoms) off San ClementeIsland, California was made with a
camerasystem conceived by Prof. John D. Isaacs(Scripps Institution
of Oceanography). Thecamera-flash lamp system is thrown over-board
and allowed to stay on the bottom un tila time delay release
mechanism permi ts thecamera to float to the surface for
retrieval.Rare bottom fish have been photographed aswell as a large
shark at a bait box on the bot-tom.
LIGHT REQUIREMENTS
Below a few hundred feet of depth in waterdaylight is
inappreciable and auxiliary light-ing is required. A lamp on or
near a camera isneeded to produce the following
beam-candela-seconds (BCPS) to expose a photo-graph.
c(BCPS) = D2A 2 - l'2D
S
where
D = lamp-subject distance and camera-sub-ject distance
A = camera lens apertures = ASA fil m speedc = a constant 15 to
25, when D is in feet
a constant 162 to 270, when D is inmeters
a = absorption co-efficient of the water innatural log units
(e=2.73)
If aD = 1 the light (BCPS) must be increasedby 2.732 , or 7.5,
compared to air use. At twoattenuation lengths the factor is 55.
Beyondthis distance the ligh t losses are so great thatphotography
is almost impossible, especiallydue to the image loss in the
low-contrastphotograph. No convenien t easy-to-use meteris
available to measure the factor, ea2D , whichcould be called the
water factor.
The absorption coefficient depends uponthe wave length (color)
of the light with the
largest value in the red. Duntleyl gives in-formation for
distilled water (Table 1). It isobserved that the red light will be
absorbedmuch more than the green or blue with dis-tance, thus
disturbing the color balance.
Suspended particles in the water of theocean absorb and scatter
the light so that theattenuation length is much shorter than
theabove information for distilled water.
CAMERA CONTROL
A very practical problem of photography inthe sea is measurement
and control of theheight of the camera over the bottom
whilephotographs are being made. A camera forexample, must be held
accurately at someselected height, such as 10 meters from
thebottom. There are at least four ways to dothis.
1. The trigger2 method where a shutteroperating device, usually
a switch, hangs be-low the camera. This bottom-activated
elec-trical swi tch causes operation of the camerawhen it is at the
desired height. After the ex-posure is made the camera is raised in
prep-aration for the next operation.
2. The use of a pinger3 sound source on thecamera which enables
the operator on the shipto know at all times where his camera
islocated above the bottom. A pinger sends outa short powerful
pulse of underwater sound atregular one second intervals. One sound
wavegoes directly to the surface from the pinger,and another goes
down to the bottom of thesea, is reflected, and then arrives at a
latertime at the surface. The operator on the shipmeasures this
time delay between the twosignals, and thereby knows where his
camerais positioned as the velocity of sound in wateris known to be
about 5,000 feet per second.
3. The sled system of Capt. Jacques Cous-teall. His device, a
special sled which carriesthe camera and strobe, has the ability to
rightitself regardless of how it is dragged across thebottom by a
strong cable from the ship.
The sled, or troika as Cousteau calls it, is anideal platform
for a deep sea camera. Thelamps and cameras are carefully placed
be-hind the heavy metal parts of the sled toreduce the probability
of damage as the sledgoes through rough areas on the bottom of
the
Color Blue
TABLE 1
Green Red
Wave length, mlJ.l/a (In) meters
40013
44022
48028
52025
56019
6005.1
6503.3
7001.7
-
SlJPPLEME TTARY LlGHTI TG IN UNDERWATER PHOTOGRAPHY 909
sea. Cousteau is always careful to make a de-tailed sonar study
of the area he plans to draghis sled over, also observing winds and
tides.Then lowering the device to the bottom, hebegins traversing
over the area of interest.
What if the sled becomes wedged into atough spot and becomes an
anchor? Cousteaupartially solves this problem with a doublecable
attachment method. The main cable isfirmly attached to the stern of
the sled. Aweaker cable ties the main cable to the bow.If the sled
is stuck, a powerful vertical pullwill break the weak connection
and the sledwill upend, hopefully releasing itself from thebottom
attachmen t.
Cousteau once dragged his sleds over themid-Atlantic rift
mountains and obtainedsome remarkable photographs. One of
theseappears as an illustration in his book TheLiving Sea'. I urge
you to take a look at thatmoon-like scene and wonder how in the
worldhis sled could go in such a rough place withouttrouble.
Incidentally, one of his sleds is still onthe bottom, about halfway
to Europe. If youwant this sled with its movie camera andstrobe for
a souvenir I am sure that Cousteauwill give it to you if you find
it. Please notethe location, the sled is halfway betweenMonaco and
New York, possibly north of thedirect line by some 500 or a
thousand miles!
4. The captive vehicle5 method. One systemuses a TV camera on a
tethered submersiblewith a cable to the master ship. A
monitorscreen permits the operator on the ship tocontrol the
submersible into the desired pat-tern. Then when the subject is in
Yiew, thephotographic equipment is turned on to ob-tain high
quality photographs for measure-men t or record purposes.
An ambitious project was undertaken in1961 on the continental
shelf off l\ewfound-land to photograph submarine cables. Asstated
by G. R. Leopold in his Bell TelephoneLaboratory report (File f O.
34912-16), "Thepurpose of the investigation was to examinethe
cables and the ocean bottom at firsthandand to seek out possible
physical explanationsfor our apparent vulnerability to
trawlerbreaks in the area."
This vehicle, with a controlled buoyancyframe using electrical
propulsion, was madeby the Vare Industries. An electric cablejoined
it to the mother ship, the Polar Star.On board the vehicle was a
television screento record the action below, and the
necessarycontrols to guide the vehicle to follow thecable once it
was found.
A deep-sea 35-m m. camera and a 100-watt-second xenon flash lamp
were mounted near
the TV camera tube as a remote view finder.An operator watching
the TV screen couldtake photographs whenever he desired bypressing
a push button.
Another very difficult but successful deep-sea effort was the
photography of theThresher site, some 200 miles east of Boston,in
8,500 feet of water. The ill-fated Th.resherwas a nuclear powered
submarine that sunkduring her deep diving tests in April, 1963.Many
groups worked on the photographyeffort. Worzel of the Lamont
Geological Ob-servatory, Columbia University, while aboardthe
Robert D. Conrad made many photographswith a Thorndike Camera of
the bottom con-tact type. Hersey used the pinger system onseveral
Edgerton, Germerhausen & Griercameras, together with four
widely spacedhydrophones while aboard the Atlantis II,for camera
location information. Buchananand Patterson, Office of Naval
Research, usedcameras with remote surface control. Thecameras were
started from the surface when amagnetometer near the camera showed
thepresence of iron. Some 40,000 photos weremade within 1,000 feet
of the wreck. Fromthese a photomosaic was made. Part of theThresher
wreck was photographed with thenumber of the submarine clearly
visible.
The bathyscaphe Trieste (remodeled andcalled Trieste II) made
eight dives at theThresher si teo Photographs were made andalso a
clearly marked piece of pipe was re-trieved from the bottom.
Excellent photographs of the H-bomb thatwas lost and recovered
off Spain were madefrom the Alvin and other underwater
vehiclesduring the past year.
There are many cameras at the bottom ofthe sea. I personally
know of one stereo pairthat is full of exposed pictures, probably
thebest that I have ever made. This loss was ex-perienced at a spot
on the north wall of thePuerto Rico Trench at 20°07' North,
66°30'\,Vest. '0le were pulling the gear up with thewinch after the
photo run. Suddenly the cableparted and down it wen t. There are
thou-sands of feet of cable on top of the camera.
I thought it would be worth a retrievaleffort, so the next year
I was at the same spotwith an improvised hook and had the use ofthe
ship for an all night dragging search. Noluck. Perhaps we snagged
the cable but itslipped free. \Ve were probably at the wrongspot
due to navigation errors. Better fishingnext time! If anyone finds
the camera systemplease remember that the film should bedeveloped.
One camera has Plus X film, theother has super Ektachrome color
film.
-
910 PHOTOGRAMMETRIC ENGINEERl NG
POSITIONING AND NAVIGATION
Bottom photography usually is uneventful,showing many miles of
desert-like sediment.Therefore, it is not too important to
knowexact positions. If one expects to photographthe identical area
for a second ti me, it will bevery difficult for two reasons: (1)
the exactshi p' s posi tion is not known; and (2) tides andcurrents
displace the camera in an unknownmanner even if the ship is held
directly abovethe target. Some method of bottom navigationis
required if the camera is to be brought backto an identical bottom
area.
One way to locate a spot on the bottom is toanchor a buoy. The
position of the buoy, withrespect to the anchor position, will
depend onthe water currents and their effects on thefloat. Tight
cables from large buoys to a largeanchor tend to minimize the
positional error.
A second and more involved method is asonar transponder on the
bottom which sendsa ping when commanded. Let us call this
asound-house, as con trasted to a lighthouse usedin surface
navigation. \iVhenever one sends asound signal from the surface, or
from apinger on a camera, he will receive an echoresponse at a
delay corresponding to the dis-tances involved. Thus, he can
measure twodistances, one to the bottom, and a second tothe
sound-house.
A transponder sound-house receives soundpulses from the sonar on
the ship and returnsa re-enforced echo which is also recorded onthe
sonar receiver. A series of signals will ap-pear on the chart from
the sound-house. Theship and sound-house are at their closest
posi-tion when the signal is received at the earliesttime on the
recorder chart. A second pass canthen be made from another
direction to ob-tain further information.
If two sound-houses are used, two positionscan be located on a
chart where the shipmight be located. With three sound-houses,there
is no ambiguity in the ship's position.The positioning and
navigation problem isexactly the same as the surface
navigationalfix where information from three known light-houses is
used. A lighthouse gives an angle; incontrast, a sound-house gives
an echo delaytime which is proportional to distance as thevelocity
of sound in water is almost a con-stant. The bottom navigation
problem is moredifficult than the surface navigation problembecause
of the added complication of waterdepth. The submerged water craft
and anaircraft have exactly the same problem ofnavigation in a
three-dimensional medium.Several problems may be experienced in
the
initial installation of the sound-houses; butonce they are in
place, an accurate survey canbe made and the ship should be able to
gorepeatedly to the same pinpoint spot in thesea.
N ow the camera, or other devices, can alsobe accurately located
by the use of a pinger onthe object lowered. The pulses from
thepinger will give the operator on the ship theusual
camera-to-bottom distance. At thesametime, the pulses from the
camera willstimulate the transponder sound-houses andthey in turn
will send additional signals to theship. These signals enable the
operator toknow where his camera is located with respectto the
target, if the position of the target isknown with respect to the
sound-houses.Then the remaining job is the manipulation ofthe ship
so that the camera photographs thedesired subject. This requires
skill. The samesonar procedure can be used with the bathy-scaphe,
with the additional feature that ahuman observer can take over the
controlswhen the subject comes into view. Also, theuse of a
side-looking6 sonar can be very help-ful if the target sticks out
of the bottom, oncethe bathyscaphe has been put into the
desiredarea.
The sound-house transponder can be usedas a marker. For example,
a submarine couldtake one along on her dives with the provisionfor
dropping it at the main center of interest.In this way the sub can
go back to the identi-cal spot for another effort to photograph
orsearch.
In conclusion, it seems evident that soundwill be the principal
way in which men andinstruments can do precise navigation in
thesea. Both light and radio are not effective forlarge distances
underwater because of absorp-tion. Sound will be effective to
several mileswhereas photography is only effective to 30meters
under ideal conditions. However, thetwo methods are used together
for obtaininginformation about the bottom of the sea; andsound, of
short pulse length6, can furthermoregive important information to
the geologyand archaeology of the sediments below thebottom.
REFERENCES
1. Duntley, S. Q.! "Light in the sea," Journat ofthe Opttcal
Soctety of America, 53, (2), pp. 214-233, February 1963 (This is a
review article withmany references).Tyler, J. E., W. H. Richardson,
and R. \71/.Holmes, "Method for Obtaining the OpticalProperties ?f
Large Bodies of Water," Jomnalof Geophysual Research, 64, (6), pp.
667-73,June 1959.
-
SUPPLE.IENTARY LIGHTING IN UNDERWATER PHOTOGRAPHY 911
Preisendorfer, R. W., "Divergence of the lightfield in optical
media," Scripps Institute ofOceanography, Univ. Calif., Ref. 58-41,
p. 16,1957.
2. Ewing, M., A. Vine, and J. L. \\'orzel, "Photog-raphy of the
ocean bottom," J. Opt. Soc. Am.. ,36, (6), pp. 307-321, June
1946Edgerton, H. E., and R. T. Troutner, "Deep seaincandescent
lamps," Under Sea Technology,pp. 21-26, April 1966Edgerton, H. E.,
and J. R. Killian, Jr., "Flash,seeing the unseen," Branford Press,
Newton,Mass. (2nd Edition) 1954
3. Edgerton, H. E., and J. Y. Cousteau, " nder-water camera
positioning by onar," Rev. Sci.Inst., 30, (12), pp. 1125-26,
December 1959Edgerton, H. E., " nderwater pinpoint pho-tography,"
SPIE Technical Symposium LosAngeles, California, AUgllst 8, 1963
(Journal ofthe SPIE, pp. 3-5, Oct./Nov. 1963)Edgerton, H. E., "
Underwater photographywith strobe lighting," and "Sound in the
sea,"
SPIE Convention, Santa Barbara, Cal. Oct.1966.Nalwalk, A. J., J.
B. Hersey, ]. S. Reitzel &H. E. Edgerton, "Improved Techniques
of deep-sea rock-dredging," Deep-Sea Research, pp. 301-302,
Aug./Sept. 1961
4. Cousteau, J. Y., "The Living Sea," Harper &Row, p. 239,
1963
5. Leopold, G. R., Bell Telephone Lab Report (FileNo.
34912-16)
6. Yules, J. A., and H. E. Edgerton, "Bottomsonar search
techniques," Undersea Technotogy,pp. 29-32, November 1964Edgerton,
H. E., and G. G. Hayward, "The'boomer' sonar source for seismic
profiling,"Journal Geophysical Research, 69, (14), pp. 3033-42,
July 15,1964
7. Hersey, J. B., "Sound Reflections In and UnderOceans,"
Physics Today, 18, (II), pp. 17-24,November 1965
8. Edgerton, H. E., "Exploring the sea withsonar," Discovery,
pp. 40-45, September 1966
(Continued from page 905)
color control in printing far more accuratethan any combination
of filters with trans-parency film can provide in the rapidly
chang-ing undersea conditions.
Left until last in this report is the mostversatile of
color-control tools when usedskillfully-artificial light. Flash
bulbs (bothclear and blue), photoflood lamps, quartziodide lamps,
and electronic flash are excel-lent agents of color restoration.
Blue flashbulbs and electronic flash approximate thecolor
temperatures of daylight. For shortlight-to-subject distances they
bring out colorwell. Longer distances reduce their effective-ness.
An ideal ligh t-to-su bject distance of 5 to6 feet gives a subtle,
pleasing effect of partialrestoration but does not overpower the
blueunderwater feeling. Flash weak enough to bea fill-in and not a
main light source is espe-cially effective at this distance.
Clear flash bulbs, or incandescent lampsburning with a warm
light, are effective colorcontrollers at greater distances than
colderdaylight-balanced light sources. The blue ofthe water acts as
a normalizing filter when thelights are used with a daylight film
instead ofthe tungsten film for which they were de-signed. vVhen
too close to the subject, how-ever, the underwater effect can be
lost.
Underwater photography is growing out ofits adolescence and
equipment is becomingmore sophisticated. Some day underwaterscenics
will show subtle color nuances underthe grey-blue mantle, and
underwater close-ups will modify the fiery reds and mustardyellows
of coral in back of startlingly unreal-looking fish to give a more
genuine feeling forthe fish's watery environment.
I should like to see the twilight world ofunderwater penetrated
but not overpowered.